List your company
ResearchFormerly WizWand

Multimodal Reasoning on VisualProcessBench Overall

54.6FEI AverageVRPRM-MiMo
24.450432.277740.10547.9323Aug 5, 2025
Updated 5mo ago

Evaluation results

MethodLinks
VRPRM-MiMo# Samples=53.6K, Explicit Reasoning (CoT)=true, RL Training=true2025.0854.672.91
VRPRM-MiMo w/o CoT# Samples=53.6K, Explicit Reasoning (CoT)=false, RL Training=true2025.0853.0371.62
VRPRM-Qwen3# Samples=53.6K, Explicit Reasoning (CoT)=true, RL Training=true2025.0851.8270.01
VRPRM-MiMo w/o RL & w/o CoT# Samples=3.6K, Explicit Reasoning (CoT)=false, RL Training=false2025.0851.6146.77
MiMo-VL-7B# Samples=unk2025.0851.0963.38
VRPRM-MiMo w/o RL# Samples=3.6K, Explicit Reasoning (CoT)=true, RL Training=false2025.0850.7760.61
VRPRM# Samples=53.6K, Explicit Reasoning (CoT)=true, RL Training=true2025.0849.7266
VRPRM-Qwen3 w/o RL# Samples=3.6K, Explicit Reasoning (CoT)=true, RL Training=false2025.0849.2457.07
Qwen3-VL-4B-Thinking# Samples=unk2025.0849.1861.4
Qwen2.5-VL-7B# Samples=unk2025.0848.3745.36
VRPRM-Qwen3 w/o RL & w/o CoT# Samples=3.6K, Explicit Reasoning (CoT)=false, RL Training=false2025.0848.353.39
VRPRM-Qwen3 w/o CoT# Samples=53.6K, Explicit Reasoning (CoT)=false, RL Training=true2025.0847.1468.63
Qwen2.5-VL-72B# Samples=unk2025.0846.5447.82
VRPRM w/o RL# Samples=3.6K, Explicit Reasoning (CoT)=true, RL Training=false2025.0845.7762.11
VRPRM w/o RL & w/o CoT# Samples=3.6K, Explicit Reasoning (CoT)=false, RL Training=false2025.0844.5852.2
InternVL2.5-8B# Samples=unk2025.0843.8946.08
Gemini-2.0-Flash# Samples=unk2025.0843.6842.17
VRPRM w/o CoT# Samples=53.6K, Explicit Reasoning (CoT)=false, RL Training=true2025.0838.1453.66
GPT-4o-mini# Samples=unk2025.0836.4535.91
VisualPRM-8B# Samples=400K2025.0825.6161.03